Interesting experiment in load matching

[Doc B.]

After an interesting phone call with a bottlehead who was experimenting with his crack and low impedance phones this afternoon I decided to emulate his experiment. The idea is very simple, you just put a 200-300 ohm resistor in series with each signal hot leg so the amp sees a higher impedance load. We just rigged up a bunch of clip leads and gave it a shot, comparing the circuit with the resistors in series vs. straight wire. We used MB Quart QP 400s (40 ohms) and AKG K701s (62 ohms)

We used a Crack with Speedball, all stock components. Shawn thought the midrange was made a little grungy or veiled with the resistors. Perhaps that could be improved with premium resistors, we were just using inexpensive metal films. I didn't notice that issue as much as Shawn, but I felt that the bass pop or punch was improved quite a bit with the resistors in series. There wasn't really more bass, but it was better defined. We both felt the Smack sounded cleaner and more natural in the midrange and treble than either configuration of the Crack but that the Crack might edge out the Smack on the very bottom.

I happen to find this to be the case with HD600s on the Crack vs. Smack too, and I find that I lean a bit towards the Crack with those cans. It might be worthwhile to repeat this test with a Crack that has a 5998 and film output caps, and I will do that when we have the time.

[ironbut]

When I was using Beyer 880's (250 ohm), I used to use a 1/4" adapter plug sometimes. The 880's are bass shy IMHO and the addition of the 120 ohm adapter did bring out more bass but it did so at the expense of a little upper mid (presence region 2-4kHz) loss and a bit of the top end "sparkle" (10-15kHz).
There used to be a couple of companies that made these adapters (mine is made by the German company Oehlbach) and I believe that most were 120 ohm. I seem to recall that there were some with higher resistance too.
I still have mine and I'll give it a try with my Audio Technica AD2000 and see if it makes a difference with my Crack this week.

[Laudanum]

Hmmm.  I dont find the 880's (250 ohm pro) to be bass shy but it would be a relative assessment I think.  I have a set of 990's as well (250 ohm pro's) and while the 990's have signifigantly more bass than the 880's (and probably most popular cans)  I still dont find the 880's bass shy at all even when going back to the 880's after listening to the 990's.  They may be shy in the lower bass but most music I listen to (mostly blues, blues rock, rock and even some classic metal) doesnt have much if any lower bass so I really cant comment there.  It's mostly bass and mid bass.
My main source is an old AH Tjoeb 99 tube (6922 based - modified inexpensive Marantz player) which I find to be on the warm side. Im using Crack with speedball, Mundorf MKP 100uf output caps and an RCA 6AS7 and currently rolling some Sylvania 6SN7's for the driver.  So it could just be a combination of my components.

[Armaegis]

I've read conflicting things, some say those resistors is adding to the output impedance and others say it adds to the load...

I've rigged up a couple cables with resistors built in, and my experience is that it reduces noise and turns down the treble a bit. It would be interesting to see someone plot a square wave function and see how it changes.

[xcortes]

It might be worthwhile to repeat this test with a Crack that has a 5998 and film output caps, and I will do that when we have the time.

I have a Crack with Speedball in the making with output film caps and a Western Electric 421A output tube. The output impedance with this tube should be about 70 Ohm instead of the 120 Ohm of the stock.

I had missed this thread but found it today after recalling that Etymotics makes two versions of the famour 4R (is that the name). One version has higher impedance and supposedly sounds better. The other has lower impedance. People have found that the difference is a resistor in the cable.

I'll compare vs a Ni MQ Sex. Maybe, just maybe I can get rid of opts and still use my Sonys.

Edit: I also have an unused 1K Penny & Giles pot. Perfect for trimming the resistance!

Edit 2: Now even Headroom sells the "converter": http://www.headphone.com/accessories/adaptors/etymotic-er-4p-to-er-4s-adapter.php

[Doc B.]

A couple of things to impose here -

Firstly, I misstated the impedance of the MBQuart QP400s. They are 300 ohms.

Secondly, last week we performed an experiment of doubling the current through the 5998 tube in the lab Crack amp. This should have reduced the output impedance considerably and we found that, indeed, our 62 ohm K701s were working much better with the modded Crack. If the mod holds up over extended play (we'll find out at RMAF, where I will be demoing with this modded Crack) we will offer a small kit that allows one to change a stock Crack to this setup. Eventually we might also offer one for the Speedball version, but it may require a new PC board with bigger heat sinks.

[xcortes]

So the quest continues!

So I may build my Crack without the speedball first.

I also found this interesting article that discusses the effects of adding resistors to the output jacks. Apparently they're not minor:

http://www.meier-audio.homepage.t-online.de/tipstricks.htm

(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fwww.meier-audio.homepage.t-online.de%2Fbilder%2Fattenuation.gif&hash=813e92162277d4c35cb7d5164d8ddcbfba9c210f)

[tdogzthmn]

Very interested in this potential mod.  I'm running my crack+speedball with the 5998 and an E80CC but I still find the HD600 to be stronger than my other two AKG cans.  If this mod is able to get lower impedance cans working better than I can put off buying a smack for now.